Publication
Traffic forecasting problems of freeway networks are heavily tackled by deep learning methods because it requires learning highly complex correlations between variables both in time and space. Adopting a graph convolutional network (GCN) becomes a standard to extract spatial correlations; therefore, most works have achieved great prediction accuracy by implanting it into their architecture. However, the conventional GCN has the drawback that receptive field size should be small, i.e., barely refers to traffic features of remote sensors, resulting in inaccurate long-term prediction. We suggest a forecasting model called two-level resolutions deep neural network (TwoResNet) that overcomes the limitation. It consists of two resolution blocks. The low-resolution block predicts traffics on a macroscopic scale, such as regional traffic changes. On the other hand, the high-resolution block predicts traffics on a microscopic scale by using GCN to extract spatial correlations, referring to the regional changes produced by the low-resolution block. This process allows the GCN to refer to the traffic features from remote sensors. As a result, TwoResNet achieves competitive prediction accuracy compared to state-of-the-art methods, especially showing excellent performance for long-term predictions.